Question 1

Why does ChatGPT always recommend the same law firms?

Accepted Answer

ChatGPT and other AI assistants draw on training data that disproportionately reflects the firms with the heaviest public presence: BigLaw brands like Skadden, Latham, and Kirkland that are cited in legal news, M&A coverage, and Supreme Court filings; and aggregator platforms like Avvo, Martindale-Hubbell, and FindLaw that have spent decades building structured attorney directories. When a user asks an AI assistant to recommend a corporate litigator or an M&A attorney, the model retrieves from this highly concentrated corpus. Mid-market and regional firms — even excellent ones with deep domain expertise — simply do not appear in the training data at sufficient density or with sufficient entity context for the model to include them. The structural cause is that most law firm websites are built for human navigation, not machine extraction. They lack structured data, their attorney bio pages are thin on demonstrable expertise, and they publish little original content that AI assistants can cite as evidence of authority. Until those information gaps are closed, the citation defaults will not change.

Question 2

What makes a law firm's website AEO-ready for AI search?

Accepted Answer

An AEO-ready law firm website has five structural properties that the average firm website lacks. First, every attorney bio page is detailed and factual — not a marketing-speak paragraph but a structured record of cases, publications, bar admissions, speaking engagements, and verified outcomes. Second, every practice area page answers the actual questions clients ask AI assistants, structured as direct answers with named subtopics. Third, the site deploys LegalService, Attorney, and Organization schema at the page level, exposing machine-readable facts about specializations, jurisdictions, and credentials. Fourth, the site publishes original substantive content — client alerts, case analyses, regulatory updates — on a consistent schedule that gives AI crawlers freshness signals. Fifth, the firm is referenced on authoritative external sources: bar association directories, legal news outlets, court filings databases, and peer-review platforms. A firm website that checks all five boxes is structurally visible to AI assistants in a way that a typical BigLaw-template marketing site is not.

Question 3

How can a small law firm compete with BigLaw in ChatGPT recommendations?

Accepted Answer

Small and mid-market law firms have one structural advantage over BigLaw in AEO: specificity. AI assistants are not good at nuance when recommending large generalist firms — they default to the names they see most often. But when a user asks a specific question — best employment attorney for wrongful termination in Denver, or who handles data breach class actions for mid-size companies — the model shifts from brand recognition to expertise matching. A 15-attorney firm that has published thirty detailed articles on Colorado employment law, maintained an attorney bio page with verified case outcomes, and earned citations in Colorado Bar Association publications can outperform a national firm on that specific query. The playbook is to dominate a narrow topic and geography intersection rather than compete at the national brand level. Start with two to three practice area and geography combinations, build genuine content depth in those intersections, and measure citation share at that specific query level. Competition is far less crowded — and the client value when the citation lands is higher.

Question 4

What schema markup should a law firm or attorney use?

Accepted Answer

Law firms and attorneys have a well-defined schema vocabulary that most firms are not using. At the organization level, the LegalService type with attorney sub-types is the correct starting point — it signals to AI crawlers that the entity is a legal service provider rather than a generic business. Each attorney should have a Person schema with jobTitle, alumniOf, memberOf (for bar associations), knowsAbout (for practice areas), and hasCredential fields populated. Practice area pages should use Service schema with serviceType, areaServed, and provider linked back to the Organization entity. FAQ content should be wrapped in FAQPage schema — this is one of the highest-value schema implementations for legal AEO, because clients ask highly specific legal questions that FAQ schema matches directly. Finally, any published legal content — client alerts, case analysis, regulatory updates — should use Article or LegalDocument schema with author attribution and datePublished. The firms seeing the highest citation rates in our 2026 audit have implemented all five schema layers. The median firm in the same sample has implemented zero.

Question 5

How do YMYL rules affect legal content in AI search?

Accepted Answer

YMYL — Your Money or Your Life — is a content classification that Google introduced for pages where inaccurate information could cause direct financial or physical harm, and it applies with full force to legal content. For AI assistants, the YMYL constraint operates similarly but with different mechanics: models are trained to apply additional caution when generating or citing content on legal topics, meaning they are more likely to recommend authoritative sources and less likely to synthesize answers from thin or promotional content. This is both a risk and an opportunity for law firms. The risk: AI assistants will hedge, disclaim, and sometimes refuse to recommend specific firms for specific legal situations if the query touches on active litigation, jurisdiction-specific advice, or highly consequential matters. The opportunity: firms that have invested in genuinely authoritative content — content that reads like it was written by an expert for an expert, cites specific statutes and case law, and acknowledges the limits of general information — get treated as credible sources by AI models precisely because YMYL caution filters out the thin, promotional content that dominates the majority of legal marketing websites. YMYL is a sorting mechanism that rewards genuine expertise.